Auditory Brainstem Responses in Congenital Heart Disease Yasuo Sunaga, MD*, Katsuhiko
Sone, MDt, Kanji Nagashima,
Takayoshi Kuroume,
To evaluate the effect of chronic hypoxemia on brainstem maturation, auditory brainstem responses were examined in 70 children (32 with and 38 without cyanosis) who had congenital heart disease. Ninety-one age-matched normal children served as controls. At 1-3 months of age, the I-V interpeak latencies of cyanotic infants (mean _+ S.D.; 5.17 + 0.17 ms) were more prolonged than were those of controls (4.95 + 0.11 ms) and those without cyanosis (4.84 + 0.22 ms; P < .05; P < .01). At 4-11 months of age, the I-V interpeak lateneies of cyanotic infants (4.85 _+0.13 ms) were more prolonged than were those of controls (4.67 _+0.19 ms) and those not experiencing cyanosis (4.5 + 0.17 ms; P < .05; P < .01). In the cyanotic children, there was a significant negative correlation between the I-V interpeak latency and oxygen partial pressure (P < .01) or oxygen saturation (P < .05). Three of the 70 patients (4.3%) with congenital heart disease had absent auditory brainstem response. These data indicate that chronic hypoxemia may be one of the factors in retarded brainstem maturation. Sunaga Y, Sone K, Nagashima K, Kuroume T. Auditory brainstem responses in congenital heart disease. Pediatr Neurol 1992;8:437-40.
Introduction Parents and physicians generally have recognized delayed developmental milestones in infants and children with congenital heart disease, especially cyanotic heart disease [ 1]. Developmental retardation appears mainly to be a result of undernutrition or chronic hypoxemia [1] affecting the cerebral cortex; however, this retardation may be partially due to delayed brainstcm maturation. There have been only a few reports detailing auditory brainstem responses (ABRs) in children with congenital heart disease [2,3]. ABRs, measured by a noninvasive and reproducible examination, have been used as a monitor for the assessment of brainstem function or brainstem maturation [4-6].
From the Department of Pediatrics; School of Medicine; Gunma University; rGunma Children's Medical Hospital; Gunma, Japan.
MD*, and
MD*
In our study, we used the ABR test to evaluate the influence of chronic hypoxemia on brainstem maturation in children with cyanotic heart disease. Methods We examined 70 children with cyanotic or acyanotic heart disease, excluding patients with cardiac syndrome complex (e.g., Down syndrome, Marfan syndrome), premature birth, asphyxia, and hyperbilirubinemia. The age of patients ranged from 1 month to 4 years. The age distribution of patients cyanotic/acyanotic was 1-3 months - 7/9; 4-11 months - 10/7; 12 months - 4 / 1 2 , and 2-4 years - 11/10. Table 1 lists the distribution of types of congenital heart disease. As a comparison with the patients with congenital heart disease, we also studied 91 normal children, ages 2 days to 4 years: 22 neonates (2-5 days of age); 18 ages 1-3 months; 20 ages 4-11 months; 8 age 1 year; and 23 ages 2-4 years. These children served as study controls. In our patients with congenital heart disease, ABR tests were performed prior to curative cardiac surgery. All patients were tested in spontaneous or sedated sleep. Electrodes were attached to the vertex (Cz), left earlobe (AI), and midline forehead (FPz). Electrical activities were recorded using a Czto-A1 electrode configuration, with amplifier filter settings of 1003,000 Hz, and FPz served as ground. We averaged 1,024 responses automatically using Nihon Kohden Neuropack 2 (MEB 5100). Stimuli were binaural rarefaction clicks delivered to an earphone at a rate of 10/s and a hearing-level intensity of 90 dB. Analysis time was set at 10 ms. All patients were tested with the stimuli decreasing in intensity by 10 dB decrements to a 60 dB level in order to confirm that each ABR component was delayed with decreasing stimuli intensity. If artifacts were detected through oscilloscope monitor, the data were excluded. Oxygen saturation (Sao2) and oxygen partial pressure (Pao2) revealed an average of 2-30 times the blood gases analysis. Those patients presenting with remarkable changes of Sao2 or Pao2 were excluded from our study. The range of Sao2/Pao2 was 35-90%/20-60 mm Hg. We performed statistical analyses using analysis of variance, unpaired Student t test, primary regression test, and Scheffe's multiple comparison test.
Results Controls were divided into 5 subgroups by age. The peak latencies of waves I, III, and V, and I-V interpeak latency are presented in Table 2. There was a trend for all peak latencies and I-V interpeak latencies to decrease with age. This trend was more pronounced for the later ABR waves.
Communications should be addressed to: Dr. Sunaga; Department of Pediatrics; School of Medicine; Gunma University; 3-39-15 Showacho, Maebashi; Gunma, 371 Japan. Received March 23, 1992; accepted June 15, 1992.
Sunaga et al: Congenital Heart Disease
437
Table l.
Distribution of types of congenital heart disease
Cyanotic Group (no. of pts)
Acyanotic Group (no. of pts)
Tetralogy of Fallot
(17)
Ventricular septal defect
(23)
Complete transposition of great arteries
(4)
Atrial septal defect
(5)
Tricuspid atresia
(4)
Patent ductus arteriosus
(5)
Double-outlet right ventricle
(3)
Pulmonary stenosis
(2)
Other
(4)
Other
(3)
The patients with congenital heart disease (with and without cyanosis) were divided into 4 subgroups by age (i.e., 1-3 months, 4-11 months, 1 year, 2-4 years), and were compared with the controls. There was no significant difference between wave I and III peak latencies among cyanotic, acyanotic, and control subjects. In the age groups, 1-3 and 4-11 months, the wave V peak latencies of cyanotic patients were more prolonged than for age-matched acyanotic patients (P < .05). In the same age groups, the I-V interpeak latencies of cyanotic patients were more prolonged than were those of agematched acyanotic patients and controls (P < .01, P .05, respectively). There was no difference in wave V peak
latencies or in I-V interpeak latencies between the cyanotic and normal neonates. In the cyanotic patients, we evaluated ~ relationship between the I-V interpeak latency and Sao2 ~w P a o 2 . There was a significant negative correlation belween 1he I-V interpeak latency and Sao2 (P < .05) or P~o7 (P < ,01: Figs 1,21. The cyanotic patients were divided into 2 groups: group A with interpeak latencies shorter than the mean + 2.0 S.D. of the age-matched controls and group B with mterpeak latencies longer than the mean + 2.0 S.D. The Sao2 and Pao2 levels of group B were significantly lower than were those of group A (P < .05 and P < .01, respectively; Table 3). Two of 32 cyanotic children and 1 of 38 acyanotic children had absent ABRs.
Discussion Our data support previous studies [5,6] revealing a decrease in peak latencies for waves I, III, V, and I-V interpeak latency with age. This change was pronounced for the later components. Furthermore, the decrease of each latency was most dramatic in infants younger than 1 year of age [4,7]. The reason for the age-distinct decrease in each wave peak latency has been believed to be progressive brainstem myelination of the auditory pathway [5]. It is well known that the myelination process appears to proceed in a cau-
Table 2. Peak latencies (ms) of waves I, 1II, V, and I-V interpeak latencies (ms) in control, acyanotic, and cyanotic groups*
Age
N
Wave III Mean S.D.
Wave V Mean S.D.
Wave I-V Mean S.D.
Newborn
Control
22
1.67
0.16
4.48
0.19
6.87
0.31
5.20
0.27
1-3 mos
Control
18
1.57
0.12
4.34
0.19
6.54
0.19
4.95
0.1 I
Acyanosis
9
1.59
0.12
4.23
0.24
6.44
0.22
4.84
(I.22
Cyanosis
7
1.60
0.20
4.32
0.20
6.76 ~
0.23
5.17~§
0.17
20
1.55
0.11
4,07
0.19
6.22
0.24
4.67
0.19
Acyanosis
6
1.56
0.06
3.91
0.27
6.06
0.19
4,50
0.17
Cyanosis
9
1.55
0.23
4.12
0.17
6.40*
0.23
4.85*§
0.13
Control
8
1.48
0.10
3.88
0.22
5.86
0.17
4.38
0.19
12
1,54
0.19
3.84
0.31
5.96
0.25
4,41
0.27
4
1.47
0.13
3.87
0.15
5.94
0.16
4.49
0.11
Control
23
1.50
0.11
3.81
0.20
5.76
0.23
4.26
(I.24
Acyanosis
10
1.50
0.25
3.76
0.24
5.74
0.33
4.23
0,19
Cyanosis
10
1.48
0.11
3.77
0.21
5.85
0.35
4.37
0.33
4-11 mos
1 yr
Control
Acyanosis Cyanosis 2-4 yrs
* The values Significant * Significant § Significant
438
Wave I Mean S.D.
are shown as the mean + S.D. Statistical significance is assessed by Scheffe's multiple comparison test. difference between cyanotic and acyanotic groups, P < .05. difference between cyanotic and acyanotic groups, P < .01. difference between cyanotic and control groups, P < .05.
PEDIATRIC NEUROLOGY
Vol. 8 No. 6
Z
5.5-1
• •
==
~, 5.0~
0
•
•
•
•
'~ 4.5-
• •
O%--
l-t--
4.030
I
I
I
50 70 Sa02 I~percent]
90
Figure 1. The relationship between oxygen saturation (Sao2) and wave I-V interpeak latencies in cyanotic patients: Y = -O.O13X + 5.59, r = 0.458, P < .05, n = 30.
dal-to-rostral direction [8]. The earlier ABR components reach plateau values before the later components; therefore, ABRs have been used to monitor brainstem maturation [4-6]. Our cyanotic patients younger than 12 months of age had more prolonged wave V peak latencies than did the acyanotic patients. Moreover, in cyanotic infants 1-3 and 4-11 months of age, the I-V interpeak latencies, reflecting brainstem conduction time of the auditory pathway, were more prolonged than were those in the age-matched acyanotic and control subjects. Although we could not measure the I-V interpeak latencies of cyanotic patients at birth, there was no difference in I-V interpeak latencies between the cyanotic patients ages 1-3 months and normal neonates. In general, circulation of the fetus with cyanotic heart disease is not accompanied by hypoxemia. Accordingly, we speculate that chronic hypoxemia in cyanotic infants retards maturation of the brainstem in extrauterine life. The retardation of
5.5]
•
±
g
brainstem maturation may be due primarily to a retarded progressive myelination process of the brainstem in the presence of chronic hypoxemia [9-11]. This retardation may have the greatest effect on infants [12-14] younger than 1 year of age with tolerance to chronic hypoxemia developing with age as evidenced by normal brainstem conduction times at 1 year of age. Moreover, in order to assess the influence of chronic hypoxemia on brainstem maturation in our cyanotic patients, we studied the relationship between I-V interpeak latency and Sao2 or Pao2. We found a negative correlation between the I-V interpeak latency and the 2 oxygen levels in all cyanotic patients. Our results confirm that the more severe the hypoxemia, the more the I-V interpeak latency is prolonged; however, age cannot be ignored as a factor affecting the I-V interpeak latency. We divided all patients with cyanotic heart disease into 2 groups: group A with interpeak latencies shorter than the mean + 2.0 S.D. of age-matched controls and group B with
o.
•
sol
•
4.5t
°o | ~ ' ~ t
•
"E 4.01
• I
10
•
I
30 Pa02
f
50 [mmHg]
Figure 2. The relationship between the oxygen partial pressure (Pao2) and wave 1-V interp e a k latencies in cyanotic subjects: Y = -0.019X + 5.46, r = 0.508, P < .01, n = 30.
Sunaga et al: Congenital Heart Disease 439
Table 3. Oxygen saturation and oxygen partial pressure of groups A and B
Oxygen saturation
Group A*
Group B**
72.3 +_ I 1.4
57.2 .+_ 13.5
P < .()5
41.4 _+9.4
30.2 _+6.6
P < .01
(%) Oxygen partial pressure (mm Hg)
References
* Group A: the I-V interpeak latencies shorter than the mean + 2.0 S.D. compared with those of age-matched controls (N = 22). ** Group B: the I-V interpeak latencies longer than the mean + 2.0 S.D. compared with those of age-matched controls (N = 8). The values are shown as the mean + S.D. Statistical significance is assessed by analysis of variance and unpaired Student t test.
interpeak latencies longer than the mean + 2.0 S.D. We evaluated the levels of Sao2 and Pao2 in these groups. Both Sao2 and Pao2 levels in group B were lower than in group A. These findings support the conclusion that the more severe the chronic hypoxemia, the more I-V interpeak latency is prolonged in all cyanotic patients. Our results indicated that chronic hypoxemia is at least one of the factors that prevents brainstem maturation of the auditory pathway. Previous reports stated that severe, prolonged hypoxia or acute anoxia induces retardation of brain maturation [9] and a delay of myelin-sheath formation in pathologic findings [10,11]. These reports may support our previously stated conclusion. Finally, 3 of 70 patients with congenital heart disease had absent ABRs (4.3%). Arnold et al. reported that there was a high incidence (16%) of hearing loss with congenital heart disease [3]. In our study, chronic hypoxemia did not affect wave I peak latency, but the cause of absent ABRs
440
PEDIATRIC NEUROLOGY
Vol. 8 No. 6
with congenital heart disease could not be determined in this study.
[1] Lindle LM, Rasof B, Dunn O. Mental developmem ~n~~'ongeni tal heart disea~. J Pediatr 1967:71:198-203. [2] Stoekard JJ. Brainstem auditory ew~ked potentials in adult and infant sleep apnea syndrome, including sudden infant death syndrome and near-miss for sudden infant death. Ann NY Acad Sci 1(~82:388: 443-65. [3] Arnold SA, Brown OE, Finitzo T. Hearing loss in chikh-en with congenital heart disease: A preliminary report. Int .I Pediatr Otorhinolaryngol 1986;11:287-93. [4] Saimny A, McKean CM. Postnatal development of human brainstem potentials during the first year of life. Electroencephalogr Clin Neurophysiol 1976;40:418-26. [5] Kaga K, Tanaka Y. Auditory brainstem response and behavioral audiometry. Arch Otolaryngol 1980;106:564-6. [6] Kaga K, Tanaka Y. The correlation of brainstem responses and behavioral audiometry in neonates, infants and adults. No To Hanatsu 1978; 10: 284-90. [7] Maurizi M, Almadori G, Cagini L, et al. Auditory brainstem responses in the full-term newborn: Changes in the first 58 hours of life. Audiology 1986;25:239-47. [8] Rorke LB, Riggs HE. Myelination of the brain in the newborn. Philadelphia: JB Lippincott, 1969; 1-106. [9] Hecox KE, Cone B. Prognostic importance of brainstem auditory evoked responses after asphyxia. Neurology 1981 ;31:1429-33. [10] Takashima S, Becket LE. Developmental neuropathology in bronchopulmonary dysplasia: Alteration of glial fibrillary acidic protein and myelination. Brain Dev 1984;6:451-7. [11] Byrue P, Welch R, Johnmn MA, Darrah J, Piper M. Serial magnetic resonance imaging in neonatal hypoxic-ischemic encephalopathy. J Pediatr 1990;117:694-700. [12] MeArdle CB, Richardson C J, Hayden CK, Nicholas DA, Amparo EG. Abnormalities of the neonatal brain: MR imaging, Radiology 1987;163:395-403. [13] Salamy A, Mendelson T. Differential development of brainstem potentials in healthy and high-risk infants. Science 1980;210: 553-5. [14] Davison AN, Dobbing J. Myelination at a vulnerable period in brain development. Br Med Bull 1966;22:40-4.
Sone, MDt, Kanji Nagashima,
Takayoshi Kuroume,
To evaluate the effect of chronic hypoxemia on brainstem maturation, auditory brainstem responses were examined in 70 children (32 with and 38 without cyanosis) who had congenital heart disease. Ninety-one age-matched normal children served as controls. At 1-3 months of age, the I-V interpeak latencies of cyanotic infants (mean _+ S.D.; 5.17 + 0.17 ms) were more prolonged than were those of controls (4.95 + 0.11 ms) and those without cyanosis (4.84 + 0.22 ms; P < .05; P < .01). At 4-11 months of age, the I-V interpeak lateneies of cyanotic infants (4.85 _+0.13 ms) were more prolonged than were those of controls (4.67 _+0.19 ms) and those not experiencing cyanosis (4.5 + 0.17 ms; P < .05; P < .01). In the cyanotic children, there was a significant negative correlation between the I-V interpeak latency and oxygen partial pressure (P < .01) or oxygen saturation (P < .05). Three of the 70 patients (4.3%) with congenital heart disease had absent auditory brainstem response. These data indicate that chronic hypoxemia may be one of the factors in retarded brainstem maturation. Sunaga Y, Sone K, Nagashima K, Kuroume T. Auditory brainstem responses in congenital heart disease. Pediatr Neurol 1992;8:437-40.
Introduction Parents and physicians generally have recognized delayed developmental milestones in infants and children with congenital heart disease, especially cyanotic heart disease [ 1]. Developmental retardation appears mainly to be a result of undernutrition or chronic hypoxemia [1] affecting the cerebral cortex; however, this retardation may be partially due to delayed brainstcm maturation. There have been only a few reports detailing auditory brainstem responses (ABRs) in children with congenital heart disease [2,3]. ABRs, measured by a noninvasive and reproducible examination, have been used as a monitor for the assessment of brainstem function or brainstem maturation [4-6].
From the Department of Pediatrics; School of Medicine; Gunma University; rGunma Children's Medical Hospital; Gunma, Japan.
MD*, and
MD*
In our study, we used the ABR test to evaluate the influence of chronic hypoxemia on brainstem maturation in children with cyanotic heart disease. Methods We examined 70 children with cyanotic or acyanotic heart disease, excluding patients with cardiac syndrome complex (e.g., Down syndrome, Marfan syndrome), premature birth, asphyxia, and hyperbilirubinemia. The age of patients ranged from 1 month to 4 years. The age distribution of patients cyanotic/acyanotic was 1-3 months - 7/9; 4-11 months - 10/7; 12 months - 4 / 1 2 , and 2-4 years - 11/10. Table 1 lists the distribution of types of congenital heart disease. As a comparison with the patients with congenital heart disease, we also studied 91 normal children, ages 2 days to 4 years: 22 neonates (2-5 days of age); 18 ages 1-3 months; 20 ages 4-11 months; 8 age 1 year; and 23 ages 2-4 years. These children served as study controls. In our patients with congenital heart disease, ABR tests were performed prior to curative cardiac surgery. All patients were tested in spontaneous or sedated sleep. Electrodes were attached to the vertex (Cz), left earlobe (AI), and midline forehead (FPz). Electrical activities were recorded using a Czto-A1 electrode configuration, with amplifier filter settings of 1003,000 Hz, and FPz served as ground. We averaged 1,024 responses automatically using Nihon Kohden Neuropack 2 (MEB 5100). Stimuli were binaural rarefaction clicks delivered to an earphone at a rate of 10/s and a hearing-level intensity of 90 dB. Analysis time was set at 10 ms. All patients were tested with the stimuli decreasing in intensity by 10 dB decrements to a 60 dB level in order to confirm that each ABR component was delayed with decreasing stimuli intensity. If artifacts were detected through oscilloscope monitor, the data were excluded. Oxygen saturation (Sao2) and oxygen partial pressure (Pao2) revealed an average of 2-30 times the blood gases analysis. Those patients presenting with remarkable changes of Sao2 or Pao2 were excluded from our study. The range of Sao2/Pao2 was 35-90%/20-60 mm Hg. We performed statistical analyses using analysis of variance, unpaired Student t test, primary regression test, and Scheffe's multiple comparison test.
Results Controls were divided into 5 subgroups by age. The peak latencies of waves I, III, and V, and I-V interpeak latency are presented in Table 2. There was a trend for all peak latencies and I-V interpeak latencies to decrease with age. This trend was more pronounced for the later ABR waves.
Communications should be addressed to: Dr. Sunaga; Department of Pediatrics; School of Medicine; Gunma University; 3-39-15 Showacho, Maebashi; Gunma, 371 Japan. Received March 23, 1992; accepted June 15, 1992.
Sunaga et al: Congenital Heart Disease
437
Table l.
Distribution of types of congenital heart disease
Cyanotic Group (no. of pts)
Acyanotic Group (no. of pts)
Tetralogy of Fallot
(17)
Ventricular septal defect
(23)
Complete transposition of great arteries
(4)
Atrial septal defect
(5)
Tricuspid atresia
(4)
Patent ductus arteriosus
(5)
Double-outlet right ventricle
(3)
Pulmonary stenosis
(2)
Other
(4)
Other
(3)
The patients with congenital heart disease (with and without cyanosis) were divided into 4 subgroups by age (i.e., 1-3 months, 4-11 months, 1 year, 2-4 years), and were compared with the controls. There was no significant difference between wave I and III peak latencies among cyanotic, acyanotic, and control subjects. In the age groups, 1-3 and 4-11 months, the wave V peak latencies of cyanotic patients were more prolonged than for age-matched acyanotic patients (P < .05). In the same age groups, the I-V interpeak latencies of cyanotic patients were more prolonged than were those of agematched acyanotic patients and controls (P < .01, P .05, respectively). There was no difference in wave V peak
latencies or in I-V interpeak latencies between the cyanotic and normal neonates. In the cyanotic patients, we evaluated ~ relationship between the I-V interpeak latency and Sao2 ~w P a o 2 . There was a significant negative correlation belween 1he I-V interpeak latency and Sao2 (P < .05) or P~o7 (P < ,01: Figs 1,21. The cyanotic patients were divided into 2 groups: group A with interpeak latencies shorter than the mean + 2.0 S.D. of the age-matched controls and group B with mterpeak latencies longer than the mean + 2.0 S.D. The Sao2 and Pao2 levels of group B were significantly lower than were those of group A (P < .05 and P < .01, respectively; Table 3). Two of 32 cyanotic children and 1 of 38 acyanotic children had absent ABRs.
Discussion Our data support previous studies [5,6] revealing a decrease in peak latencies for waves I, III, V, and I-V interpeak latency with age. This change was pronounced for the later components. Furthermore, the decrease of each latency was most dramatic in infants younger than 1 year of age [4,7]. The reason for the age-distinct decrease in each wave peak latency has been believed to be progressive brainstem myelination of the auditory pathway [5]. It is well known that the myelination process appears to proceed in a cau-
Table 2. Peak latencies (ms) of waves I, 1II, V, and I-V interpeak latencies (ms) in control, acyanotic, and cyanotic groups*
Age
N
Wave III Mean S.D.
Wave V Mean S.D.
Wave I-V Mean S.D.
Newborn
Control
22
1.67
0.16
4.48
0.19
6.87
0.31
5.20
0.27
1-3 mos
Control
18
1.57
0.12
4.34
0.19
6.54
0.19
4.95
0.1 I
Acyanosis
9
1.59
0.12
4.23
0.24
6.44
0.22
4.84
(I.22
Cyanosis
7
1.60
0.20
4.32
0.20
6.76 ~
0.23
5.17~§
0.17
20
1.55
0.11
4,07
0.19
6.22
0.24
4.67
0.19
Acyanosis
6
1.56
0.06
3.91
0.27
6.06
0.19
4,50
0.17
Cyanosis
9
1.55
0.23
4.12
0.17
6.40*
0.23
4.85*§
0.13
Control
8
1.48
0.10
3.88
0.22
5.86
0.17
4.38
0.19
12
1,54
0.19
3.84
0.31
5.96
0.25
4,41
0.27
4
1.47
0.13
3.87
0.15
5.94
0.16
4.49
0.11
Control
23
1.50
0.11
3.81
0.20
5.76
0.23
4.26
(I.24
Acyanosis
10
1.50
0.25
3.76
0.24
5.74
0.33
4.23
0,19
Cyanosis
10
1.48
0.11
3.77
0.21
5.85
0.35
4.37
0.33
4-11 mos
1 yr
Control
Acyanosis Cyanosis 2-4 yrs
* The values Significant * Significant § Significant
438
Wave I Mean S.D.
are shown as the mean + S.D. Statistical significance is assessed by Scheffe's multiple comparison test. difference between cyanotic and acyanotic groups, P < .05. difference between cyanotic and acyanotic groups, P < .01. difference between cyanotic and control groups, P < .05.
PEDIATRIC NEUROLOGY
Vol. 8 No. 6
Z
5.5-1
• •
==
~, 5.0~
0
•
•
•
•
'~ 4.5-
• •
O%--
l-t--
4.030
I
I
I
50 70 Sa02 I~percent]
90
Figure 1. The relationship between oxygen saturation (Sao2) and wave I-V interpeak latencies in cyanotic patients: Y = -O.O13X + 5.59, r = 0.458, P < .05, n = 30.
dal-to-rostral direction [8]. The earlier ABR components reach plateau values before the later components; therefore, ABRs have been used to monitor brainstem maturation [4-6]. Our cyanotic patients younger than 12 months of age had more prolonged wave V peak latencies than did the acyanotic patients. Moreover, in cyanotic infants 1-3 and 4-11 months of age, the I-V interpeak latencies, reflecting brainstem conduction time of the auditory pathway, were more prolonged than were those in the age-matched acyanotic and control subjects. Although we could not measure the I-V interpeak latencies of cyanotic patients at birth, there was no difference in I-V interpeak latencies between the cyanotic patients ages 1-3 months and normal neonates. In general, circulation of the fetus with cyanotic heart disease is not accompanied by hypoxemia. Accordingly, we speculate that chronic hypoxemia in cyanotic infants retards maturation of the brainstem in extrauterine life. The retardation of
5.5]
•
±
g
brainstem maturation may be due primarily to a retarded progressive myelination process of the brainstem in the presence of chronic hypoxemia [9-11]. This retardation may have the greatest effect on infants [12-14] younger than 1 year of age with tolerance to chronic hypoxemia developing with age as evidenced by normal brainstem conduction times at 1 year of age. Moreover, in order to assess the influence of chronic hypoxemia on brainstem maturation in our cyanotic patients, we studied the relationship between I-V interpeak latency and Sao2 or Pao2. We found a negative correlation between the I-V interpeak latency and the 2 oxygen levels in all cyanotic patients. Our results confirm that the more severe the hypoxemia, the more the I-V interpeak latency is prolonged; however, age cannot be ignored as a factor affecting the I-V interpeak latency. We divided all patients with cyanotic heart disease into 2 groups: group A with interpeak latencies shorter than the mean + 2.0 S.D. of age-matched controls and group B with
o.
•
sol
•
4.5t
°o | ~ ' ~ t
•
"E 4.01
• I
10
•
I
30 Pa02
f
50 [mmHg]
Figure 2. The relationship between the oxygen partial pressure (Pao2) and wave 1-V interp e a k latencies in cyanotic subjects: Y = -0.019X + 5.46, r = 0.508, P < .01, n = 30.
Sunaga et al: Congenital Heart Disease 439
Table 3. Oxygen saturation and oxygen partial pressure of groups A and B
Oxygen saturation
Group A*
Group B**
72.3 +_ I 1.4
57.2 .+_ 13.5
P < .()5
41.4 _+9.4
30.2 _+6.6
P < .01
(%) Oxygen partial pressure (mm Hg)
References
* Group A: the I-V interpeak latencies shorter than the mean + 2.0 S.D. compared with those of age-matched controls (N = 22). ** Group B: the I-V interpeak latencies longer than the mean + 2.0 S.D. compared with those of age-matched controls (N = 8). The values are shown as the mean + S.D. Statistical significance is assessed by analysis of variance and unpaired Student t test.
interpeak latencies longer than the mean + 2.0 S.D. We evaluated the levels of Sao2 and Pao2 in these groups. Both Sao2 and Pao2 levels in group B were lower than in group A. These findings support the conclusion that the more severe the chronic hypoxemia, the more I-V interpeak latency is prolonged in all cyanotic patients. Our results indicated that chronic hypoxemia is at least one of the factors that prevents brainstem maturation of the auditory pathway. Previous reports stated that severe, prolonged hypoxia or acute anoxia induces retardation of brain maturation [9] and a delay of myelin-sheath formation in pathologic findings [10,11]. These reports may support our previously stated conclusion. Finally, 3 of 70 patients with congenital heart disease had absent ABRs (4.3%). Arnold et al. reported that there was a high incidence (16%) of hearing loss with congenital heart disease [3]. In our study, chronic hypoxemia did not affect wave I peak latency, but the cause of absent ABRs
440
PEDIATRIC NEUROLOGY
Vol. 8 No. 6
with congenital heart disease could not be determined in this study.
[1] Lindle LM, Rasof B, Dunn O. Mental developmem ~n~~'ongeni tal heart disea~. J Pediatr 1967:71:198-203. [2] Stoekard JJ. Brainstem auditory ew~ked potentials in adult and infant sleep apnea syndrome, including sudden infant death syndrome and near-miss for sudden infant death. Ann NY Acad Sci 1(~82:388: 443-65. [3] Arnold SA, Brown OE, Finitzo T. Hearing loss in chikh-en with congenital heart disease: A preliminary report. Int .I Pediatr Otorhinolaryngol 1986;11:287-93. [4] Saimny A, McKean CM. Postnatal development of human brainstem potentials during the first year of life. Electroencephalogr Clin Neurophysiol 1976;40:418-26. [5] Kaga K, Tanaka Y. Auditory brainstem response and behavioral audiometry. Arch Otolaryngol 1980;106:564-6. [6] Kaga K, Tanaka Y. The correlation of brainstem responses and behavioral audiometry in neonates, infants and adults. No To Hanatsu 1978; 10: 284-90. [7] Maurizi M, Almadori G, Cagini L, et al. Auditory brainstem responses in the full-term newborn: Changes in the first 58 hours of life. Audiology 1986;25:239-47. [8] Rorke LB, Riggs HE. Myelination of the brain in the newborn. Philadelphia: JB Lippincott, 1969; 1-106. [9] Hecox KE, Cone B. Prognostic importance of brainstem auditory evoked responses after asphyxia. Neurology 1981 ;31:1429-33. [10] Takashima S, Becket LE. Developmental neuropathology in bronchopulmonary dysplasia: Alteration of glial fibrillary acidic protein and myelination. Brain Dev 1984;6:451-7. [11] Byrue P, Welch R, Johnmn MA, Darrah J, Piper M. Serial magnetic resonance imaging in neonatal hypoxic-ischemic encephalopathy. J Pediatr 1990;117:694-700. [12] MeArdle CB, Richardson C J, Hayden CK, Nicholas DA, Amparo EG. Abnormalities of the neonatal brain: MR imaging, Radiology 1987;163:395-403. [13] Salamy A, Mendelson T. Differential development of brainstem potentials in healthy and high-risk infants. Science 1980;210: 553-5. [14] Davison AN, Dobbing J. Myelination at a vulnerable period in brain development. Br Med Bull 1966;22:40-4.
